Algorithm 1 Adaptive Gabor filter and image feature fusion.

Input: Original image of finger vein: I

Step 1: Extract the region of interest of the image I, add 0 to the edge of the image to transform its size to 256 × 256.

Step 2: Divide the image into 16 × 16 sub-blocks and then obtain the gabor function window width of each sub-block.

for each row u∈1, 2, …, 16 do

for each column v∈1, 2, …, 16 do

Compute the gradient components in X and Y directions per Equation (3).

Obtain the window width of Gabor filter per Equation 4.

end

end

Step 3: Gabor transform for each sub-block image:$I_n^{\text{'}}=g_n\left(x,y\right)*I_n\left(x,y\right),$ and obtain Gabor transform of the whole image.

Step 4: Fuse the Gabor image features in 8 orientations with the same gabor transform size per Equation (3).

Output: Finger vein feature fusion image: $F_u\left(z\right).$